IEEE - Institute of Electrical and Electronics Engineers, Inc. - AI2: Safety and Robustness Certification of Neural Networks with Abstract Interpretation

2018 IEEE Symposium on Security and Privacy (SP)

Author(s): Timon Gehr ; Matthew Mirman ; Dana Drachsler-Cohen ; Petar Tsankov ; Swarat Chaudhuri ; Martin Vechev
Publisher: IEEE - Institute of Electrical and Electronics Engineers, Inc.
Publication Date: 1 May 2018
Conference Location: San Francisco, CA, USA
Conference Date: 20 May 2018
Page(s): 3 - 18
ISBN (Electronic): 978-1-5386-4353-2
ISSN (Electronic): 2375-1207
DOI: 10.1109/SP.2018.00058
Regular:

We present AI2, the first sound and scalable analyzer for deep neural networks. Based on overapproximation, AI2 can automatically prove safety properties (e.g., robustness)... View More

Advertisement